Bias correction in maximum likelihood logistic regression

Bias correction in maximum likelihood logistic regression

This paper provides an expression for bias of the maximum likelihood logistic regression estimates for use with small sample sizes. This bias correction is based on an expansion of the maximum likelihood equation. Simulation results show these corrections to be highly effective in small samples.

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Statistics in medicine 1983-01, Vol.2 (1), p.71-78
1. Verfasser: Schaefer, Robert L.
Format: Artikel
Sprache:eng
Schlagworte:
Bias
Likelihood
Logistic
Maximum
Probability
Regression
Regression Analysis
Risk
Simulation
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper provides an expression for bias of the maximum likelihood logistic regression estimates for use with small sample sizes. This bias correction is based on an expansion of the maximum likelihood equation. Simulation results show these corrections to be highly effective in small samples.
ISSN:0277-6715
1097-0258
DOI:10.1002/sim.4780020108